← Back to series HSCI 410

Applied Biostatistics

Office Hours — the companion podcast for HSCI 410

Each episode is a relaxed walkthrough of the week's lesson — audio, summary, and full transcript on every page.

Foundations & descriptive analysis

Lessons 01 — 02
L · 01
A Structured Approach to Data AnalysisCausal diagrams, data-collection sheets, coding and entry, file and variable management, and program-mode versus interactive workflows.
Open podcast
L · 02
Data Cleaning & Descriptive AnalysesData quality assessment, cleaning strategies, handling missing data, descriptive statistics, and visualization for epidemiologic datasets.
Open podcast

Regression models

Lessons 03 — 04
L · 03 · Part 1
Linear & Logistic Regression: Linear RegressionRegression analysis, hypothesis testing, X-variable coding, collinearity detection, and interaction effects.
Open podcast
L · 03 · Part 2
Linear & Logistic Regression: Logistic RegressionBinary outcomes, odds ratios, maximum likelihood estimation, goodness-of-fit, and ROC analysis.
Open podcast
L · 03 · Part 3
Linear & Logistic Regression: Model-Building StrategiesPurposeful selection, change-in-estimate, stepwise procedures, and multi-level model building for epidemiologic research.
Open podcast
L · 04 · Part 1
Generalized Linear Models: Ordinal & Multinomial ModelsProportional odds models, multinomial logistic regression, and methods for multi-category outcomes.
Open podcast
L · 04 · Part 2
Generalized Linear Models: Count & Rate DataPoisson regression, overdispersion, negative binomial models, and zero-adjusted count models.
Open podcast

Time-to-event & dependent data

Lessons 05 — 06
L · 05
Survival DataKaplan-Meier estimation, Cox proportional hazards, parametric survival models, and frailty models.
Open podcast
L · 06 · Part 1
Modelling Dependent Data: Introduction to Clustered DataHierarchical data structures, ICC, design effects, and methods for handling clustering in epidemiologic analyses.
Open podcast
L · 06 · Part 2
Modelling Dependent Data: Mixed Models for Continuous DataRandom intercepts, random slopes, contextual effects, REML estimation, and model diagnostics.
Open podcast
L · 06 · Part 3
Modelling Dependent Data: Mixed Models for Discrete DataGLMMs, logistic and Poisson random effects models, SS vs PA interpretation, and estimation methods.
Open podcast
L · 06 · Part 4
Modelling Dependent Data: Repeated Measures DataRepeated measures data structures, descriptive and graphical exploration, univariate and multivariate analytic approaches, and growth curve modelling for longitudinal health research.
Open podcast